Error Type

Use research rigor to avoid false claim errors, and research sensitivity to avoid nil result errors

The worst result in research is not negative data (data that contradicts theory), but data that shows nothing at all (nil results). Good research can give negative results, but nil results suggest error "noise" drowned out the effect, e. g. an unfocused RQ, a method that confounds constructs, or wrong analysis. In research:

Research must be rigorous enough to avoid false results, but still sensitive enough to gain true results. Improve research sensitivity by methods that enhance responses (e. g. motivating subjects with a competition), methods that reduce subject error (like subject training or clearer questions), and more powerful statistical methods. Unfortunately reducing one error type tends to increase the other, so reducing errors of commission to zero will increase errors of omission to 100%.


Tags: Valid, Results

Example(s)

(Use a descriptive name, e. g. "ITExample". Or click on an existing collection and edit it.)

Element/ErrorType (last edited 2008-11-13 16:28:06 by GuyKloss)

This content is licensed under a Creative Commons Attribution-Noncommercial-Share Alike 3.0 License (New Zealand)